For each testing condition, phase change was investigated at the cell level using all available cells (no exclusions) or all available cells that could be referenced to pyramidal-layer theta (All pyr. layer-referenced). For each of these datasets, four analyses are presented using different cell selection criteria. Cell selection becomes more conservative according to the following: (1) minimum number of spikes fired during a trial; (2) minimum theta-modulation (Rayleigh p threshold p = 0.05); 3) whether spike and theta-modulation criteria are met in either one of T3 and T4 (independent samples) or in both trials T3 and T4 (paired samples). The overall distribution of the preferred theta -phases of cells were strongly clustered around the “&lquote;mean preferred phase” in all conditions (Rayleigh test ps < 0.001) except in the probe trial (T4) in the NOV + SCOP condition where, relative to the other three testing conditions, some distributions were not that far below the significance threshold for the Rayleigh test for nonuniformity (asterisks). For paired samples, both the nonparametric paired-samples Moore test and the parametric Hotelling test were applied, with similar results.

↵^aAll p values for T4 Rayleigh test indicating reliability of mean preferred phase in T4 were p < 0.001 except *** p ≤ 0.001, ** p < 0.01, * p < 0.05.